TL;DR
- Kraken CLI is an open-source, single-binary execution engine developed by Kraken, providing AI agents and developers with direct and native access to the cryptocurrency markets— enabling spot trading, futures, staking, subaccount transfers, and WebSocket streaming through 134 commands, all executed via a zero-dependency Rust binary that outputs clean NDJSON for machine consumption.
- The Kraken CLI features a built-in MCP (Model Context Protocol) server, making it the first crypto exchange CLI seamlessly compatible with agent-based coding tools such as Claude Code, Codex, and Cursor, as well as terminal-native environments like OpenCode and OpenClaw— empowering AI agents to quickly comprehend and perform exchange operations without the need for custom API wrappers, nonce handling, or HMAC signing.
- Included with Kraken CLI is a local paper trading engine that allows AI agents to rigorously test trading strategies against real market data without any financial risk— operating a local state engine to track simulated balances, execute limit and market orders using live ticker data, and compute unrealized P&L entirely offline.
Built for the Way AI Agents Function
In pursuit of an agent-driven future, we are excited to present the Kraken CLI: a powerful, single-binary execution engine specifically designed to provide AI models (and developers) secure, native, and immediate access to cryptocurrency markets.
Whether you’re an algorithmic trader, a terminal expert, or a developer crafting LLM-powered financial applications, the Kraken CLI effectively bridges the gap between AI capabilities and real-time market execution.
Why We Created It: Progressing Beyond Basic APIs
APIs are tailored for systems, but when AI agents are tasked with interacting with a traditional crypto exchange API, they encounter challenges. They must create custom Python scripts, manage cryptographic nonces, handle complex HMAC-SHA512 payload signing, and devise their own retry mechanisms for rate limits.
This approach is fraught with potential errors and consumes valuable context window tokens.
We developed the Kraken CLI to eliminate this boilerplate. Written entirely in Rust, it is a zero-dependency binary that manages the infrastructure necessary for exchange connectivity, allowing you (and your agents) to concentrate solely on trading strategies.
Essential Features for Developers
Here’s what you can expect under the hood:
Integrated MCP Server
The CLI incorporates native support for the Model Context Protocol (MCP). By executing kraken mcp, the CLI transforms into a secure, self-describing plugin for agentic coding tools such as Claude Code, Codex, and Cursor, along with terminal-native environments like OpenCode and OpenClaw. This allows your AI agent to immediately grasp the schemas for all 134 available commands.
Local Paper Trading Engine
Carefully evaluate your agent’s trading logic against real market data without any financial risk. The Kraken paper suite runs a local state engine that monitors simulated balances, executes limit and market orders using live public ticker data, and computes unrealized P&L completely offline.
Universal Shell Compatibility
If your agent does not support MCP, it can utilize the terminal effortlessly. The CLI produces clean, machine-readable NDJSON output (-o json) across all commands, making it simple to pipe data into jq or input it directly back into an LLM’s context window.
Comprehensive Exchange Interface
This is more than just a fundamental price ticker. The CLI supports spot trading, futures trading, staking/earn allocations, subaccount transfers, and WebSocket streaming.
Enterprise-Grade Reliability
Integrated spot counter/decay and futures token-bucket rate limiting ensure your automated workflows remain stable, even when encountering exchange limits.
Easy One-Line Installation
curl --proto '=https' --tlsv1.2 -LsSf https://github.com/krakenfx/kraken-cli/releases/latest/download/kraken-cli-installer.sh | sh
The Agentic Future
The Kraken CLI represents the direction in which the industry is heading: a realm where AI agents are integral participants in financial markets, rather than mere afterthoughts added to outdated infrastructures. By open-sourcing this tool, we encourage developers, algorithmic traders, and AI innovators to help shape this vision and start building the future today.
Explore the complete documentation, source code, and agent prompt examples on GitHub:
Engaging in derivatives and other financial instruments, including leveraged products, carries significant risks and may not be suitable for all investors. Please refer to our Risk Disclosure for more information. Geographic restrictions may apply.
